Reductive Disproportionation of CO2 with Bulky Divalent Samarium

Oct 20, 2017 - disproportionation of CO2 in both cases with release of CO. This contrasts with the formation of the oxalate-bridged samarium dimer rep...
0 downloads 0 Views 2MB Size
Article Cite This: Organometallics XXXX, XXX, XXX-XXX

pubs.acs.org/Organometallics

Reductive Disproportionation of CO2 with Bulky Divalent Samarium Complexes Mathieu Xémard,† Violaine Goudy,† Augustin Braun,† Maxime Tricoire,† Marie Cordier,† Louis Ricard,† Ludovic Castro,‡ Elisa Louyriac,‡ Christos E. Kefalidis,‡ Carine Clavaguéra,§ Laurent Maron,‡ and Grégory Nocton*,† †

LCM, CNRS, Ecole polytechnique, Université Paris-Saclay, Route de Saclay, 91128 Palaiseau Cedex, France LPCNO, UMR 5215, Université de Toulouse-CNRS, INSA, UPS, Toulouse, France § Laboratoire de Chimie Physique, CNRS-Université Paris-Sud, Université Paris-Saclay, 15 avenue Jean Perrin, 91405 Orsay Cedex, France ‡

S Supporting Information *

ABSTRACT: The base-free divalent samarium complex Cptt2Sm (1; Cptt = 1,3-(tBu)2(C5H3)) has been synthesized in diethyl ether by salt metathesis of SmI2. Crystals of 1 suitable for X-ray study have been obtained by sublimation at 116 °C under reduced pressure. The dissolution of 1 in thf and pyridine solution leads to the solvent adducts Cptt2Sm(thf)2 (3) and Cptt2Sm(py) (4), respectively, while drying 3 under reduced pressure yields CpttSm(thf) (5). The reaction of CO2 with the base-free divalent samarium complexes Cptt2Sm (1) and Cpttt2Sm (2; Cpttt =1,2,4-(tBu)3(C5H2)) leads to the clean formation of bridged carbonate samarium dimers [Cpttt2Sm]2(μ-CO3) (7) and [Cptt2Sm]2(μ-CO3) (8). This is indicative of the reductive disproportionation of CO2 in both cases with release of CO. This contrasts with the formation of the oxalate-bridged samarium dimer reported from the reaction of CO2 with the Cp*2Sm(thf)2 complex. Otherwise, the reaction with CO does not proceed with the bulky complexes, while traces of O2 have led to the formation of the original bridged peroxo samarium dimer [Cpttt2Sm]2(μ-O2) (6). The mechanism for these reactions is studied herein by experiments and also by theoretical computations. The key result is that the different pathways are rather close in energy, which also explains why the nature of the final product, if only one is present, is difficult to predict a priori in this chemistry.



complex.10 At the opposite of this reaction, Gardiner reported in 2006 a reductive disproportionation of CO2 with a macrocyclic organosamarium(II) complex, yielding a carbonate complex and CO.11 The clean direct radical−radical coupling of CO2 to form oxalate is rare with divalent lanthanides.12 However, Mazzanti recently demonstrated that both oxalate and carbonate products can be formed from CO2 with a divalent ytterbium complex featuring the electron-rich tris(tertbutoxy)siloxide ligand.13 Cloke made a similar report with the low-valent (COT R )(Cp)*U(thf) x complex 14 (COT R = C8H6(1,4-SiMe3)2), as did Mills with a ThIII complex.15 In addition to oxalate and carbonate reaction products, Cloke also noted the formation of a bridged oxo dimer when the steric bulk and the reaction conditions (such as the temperature) are modulated.14 This also enabled envisaging the formation of the oxalate complex as the kinetic product and of the carbonate as the thermodynamic product.16

INTRODUCTION The activation of small molecules by low-valent metal complexes under mild conditions is an important topic from both societal and academic perspectives.1,2 The direct use of greenhouse gases, such as nitrous oxide,3 methane,4,5 carbon monoxide,6,7 and carbon dioxide,8 in the synthesis of fine chemicals is very appealing in the context of overall global warming and of depletion of carbon-based feedstocks. However, such reactions are difficult and, when they occur, the selectivity is a great concern.9 In this matter, the problem becomes challenging and interesting for academics, among them organometallic chemists, since the fine design and synthesis of adapted metal complexes is crucial in the development and the understanding of these reactions. In the specific case of CO2, the problem arises from the formation of multiple products whose relevance in the chemical industry is variable. The CO2 chemistry with low-valent f elements, such as divalent lanthanides and trivalent uranium complexes, is symptomatic of this selectivity problem. In 1998, Evans reported the reaction of the well-studied Cp*2Sm(thf)2 complex with CO2 to yield a very original bridged oxalate dimer © XXXX American Chemical Society

Special Issue: Organometallic Actinide and Lanthanide Chemistry Received: August 18, 2017

A

DOI: 10.1021/acs.organomet.7b00630 Organometallics XXXX, XXX, XXX−XXX

Article

Organometallics

herein insist on the subtlety in the design of lanthanide complexes for selective small-molecule activation.

The formation of oxo complexes from CO2 was reported by several groups with uranium complexes. Mazzanti reported the remarkable formation of a mononuclear uranium(V) oxo complex in a two-electron oxidation assisted by potassium.17 In 2005, Meyer described the reaction of a trivalent uranium tripodal (tris)aryloxide complex with CO2, yielding cleanly the bridged oxo complex and CO while CO reacts further with the parent complex to yield a bridge end-on CO dimer.18 In this appealing chemistry, the role of the bulkiness of the ligand in the reactivity of CO2 was highlighted by the formation of a mononuclear CO2 radical complex using the adamantyl rather than the tert-butyl substituent on the aryloxide group.19 To our knowledge, the formation of such bridged oxo complexes from CO2 has not been reported with divalent lanthanides. However, these can be synthesized from N2O.20 After the report of Meyer taking advantage of the functionalization of the uranium oxo complexes to form various novel adducts,21 a recent contribution from Roesky intends the same type of reactivity on the oxo-bridged samarium dimer of Cp*2Sm22,23 following the reaction described by Castro et al. from [Cp*2Sm]2(μ-O) and CO2.24 The reason why such reactivity is relevant to the CO2 chemistry with divalent lanthanides lies in the mechanistic studies: either oxo is the intermediate in the formation of carbonate species or the mechanism involves a bridged bent CO22− intermediate that further reacts with an additional molecule of CO2.24 Since from preceding reports it seems that the bulkiness of the ligand can modify the relative energy of these three possible pathways, (i) oxalate formation via a direct radical−radical coupling, (ii) carbonate formation through the intermediate oxo species, and (iii) carbonate formation through the bent CO22− intermediate, we thought it would be useful to test the CO2 reaction with various samarium complexes featuring steric hindrance different from that of the Cp*2Sm(thf)2 complex. Thus, in this paper, we report the reaction of CO2 with divalent samarium complexes containing cyclopentadienyl groups substituted with bulky tert-butyl groups: i.e., Cptt = 1,3-(tBu)2Cp and Cpttt = 1,2,4-(tBu)3Cp (Scheme 1). In order



RESULTS AND DISCUSSION Synthesis and Solid-State and Solution Structures of the Divalent Complexes. The synthesis of the Cpttt2Sm complex was described in an earlier work.25 The latter was synthesized in diethyl ether and sublimed. Although N-aromatic heterocycles are reduced by the Cp*2Sm(thf)2 complex, the bulkiness of the Cpttt ligands allowed simple coordination interactions with solvents or simple ligands. No electron transfer has been observed, although the redox potential was measured to be almost identical with that of Cp*2Sm(thf)2 in thf.25 Therefore, we became interested in the modulation of the bulkiness of the ligand and went back to the Cptt ligand. The ligand was first synthesized in the late 1950s26−29 and has been used intensely since then with several different applications.30−32 In organometallic f-element chemistry, this ligand has been successfully used with various elements, such as Ce,33−35 Yb,36−41 Tm,42,43 Sm,31,33,44,45 Th,32,46−49 and U,50 among others.51 Schumann41 and Andersen38,40 reported the ytterbium ether adduct and the base-free complex38 of the bis(tert-butylcyclopentadienyl) ligand, while the Cptt2Sm(thf) complex was reported in 1990.45 However, since we are interested in small-molecule interactions, we thought that a solvent-free Cptt2Sm molecule would be very useful and would avoid the coordination competition between the solvent and the ligand. Therefore, KCptt was dissolved in ether in the presence of SmI2 at room temperature. The reaction mixture was stirred for several hours and yielded a green solution. After filtration and evaporation of the solvent, a green solid of the Cptt2Sm(OEt2) complex was obtained. No crystal structure has been obtained. The base-free complex Cptt2Sm (1) can be obtained as crystals suitable for X-ray study by sublimation under reduced pressure between 114 and 116 °C. However, the sublimation leads to a very poor yield, since decomposition occurs at about the same temperature. An alternative way of producing the base-free complex is to progressively remove the Et2O by repeated dissolution of Cp tt 2 Sm(OEt 2 ) in toluene followed by evaporation of the brown solution under reduced pressure at 70 °C. The yield is then increased, but the material is not crystalline. The crystal structure of 1 was determined from the small green crystals at 150 K (Figure 1). Unlike the Cptt2Yb complex, which only exhibits intramolecular short-contact bonds, 1 features intermolecular short-contact interactions of 3.154 Å between a carbon atom of one Cp ring pointing toward the Sm center of another molecule. These interactions are longer than those reported in Cptt2Yb of 3.078(4) Å,38 in good agreement with the lanthanide contraction. Similar interactions were reported with other base-free Sm and Yb complexes.38 This has two main effects on the metric parameters of 1. The first is a dissymmetry between the two Sm−Cp distances. The one Cp in interaction with two samarium centers causes a larger Sm− Cp distance of 2.63 Å, while the other is 2.56 Å. These are significantly longer than those in the reported trivalent samarocene, ranging from 2.46 to 2.51 Å.31,33,44 The shortest distance compares well with that reported in the Cptt2Sm(thf) complex of 2.55 Å, in agreement with SmII formulation. The second effect is the increased bending of the metallocene with a Cp−Sm−Cp angle of 132.7° in comparison to that of the Cptt2Yb complex of 147°38 or to the Cpttt2Sm complex with an

Scheme 1. Reaction of CO2 with Base-Free Samarocene Complexes

to avoid side reactions and additional steric contribution of eventual solvents, the reactions with CO and CO2 have been performed from the base-free samarium complexes. Both complexes allow the formation of two carbonate complexes resulting from reductive disproportionation of CO2, as verified by experimental and theoretical mechanistic studies. The results B

DOI: 10.1021/acs.organomet.7b00630 Organometallics XXXX, XXX, XXX−XXX

Article

Organometallics

coordination number. The Cp−Sm−Cp angle is 137.9°, again in agreement with increased coordination distances that favor better linearity. In pyridine, the solution of 1 turns black and cooling this solution leads to the crystallization of the mono-pyridine adduct Cptt2Sm(py) (4). This structure is remarkable in the sense that the pyridine ring is not perpendicular to the Cp(ctr)−Sm−Cp(ctr) plane but parallel. This is relatively surprising but is reminiscent of the coordination compounds reported with Cpttt2Sm.25 The average Cp−Sm distance is shorter than that of the bis-thf adduct but is close to that of the mono-thf adduct. The Sm−N distance is longer to that of the Sm−O distance. This may be due to a larger steric bulk due to the unusual coordination. We were intrigued by the different stoichiometries obtained for the different solvent adducts. In thf, it is clear that two molecules fit around the samarium center, while in pyridine, only one is present. A titration with substoichiometric and stoichiometric amounts of pyridine was effected in toluene. The chemical shifts of the coordinated pyridine were monitored by 1 H NMR and lead to the diagram shown in Figure S3 in the Supporting Information. Since the free pyridine molecules and the coordinated pyridine exchange rapidly, only one set of signals can be followed. The observation of the slopes on the diagram indicates two different regimes that cross at about 1 equiv of pyridine, explaining why the molecule crystallizes this way. The reasons for these differences between pyridine and thf are not clear but indicate an important role of the dynamics in solution for samarium complexes with the Cptt ligand. Witness to this dynamic solution behavior is the easy desolvation of 3 at room temperature to yield Cptt2Sm(thf) (5), already reported in the literature,45 as is a similar complex of Eu, CpttEu(thf).53 Crystals suitable for X-ray study were analyzed by X-ray crystallography, and the unit cell was different from that already reported (Figure 3). Therefore, the data were recorded at 150 K for 5. The structure is essentially similar (within the esds) to that reported except that the cell contains two slightly different molecules in the unit cell. The important message gathered by the synthesis of the pyridine and thf adduct is that the solid-state structure may be misleading in comparison to the solution structure: viz., the coordinating solvent may intervene in the chemical reactivity of the divalent lanthanide fragment from both kinetic and thermodynamic considerations. The base-free material is therefore crucial for small-molecule activation studies in this area. Reaction of the Bulky Divalent Complexes with CO and CO2. The reaction with CO and Cptt2Sm (1) Cpttt2Sm (2) was investigated in deuterated toluene. The purple solutions do not change color with 1 atm of CO, and no change in 1H NMR has been observed. When a lower quality of CO was used, the purple solution of 1 faded over several hours. The 1H NMR data were recorded and were indicative of the appearance of a new compound. After one night at room temperature, the solution turned yellow. Yellow crystals were obtained at −35 °C from this solution. The structure shows a dimer of two Cpttt2Sm fragments bridged by a rare side-on peroxo (O22−) ligand, [Cpttt2Sm]2(μ-O2) (6) (Figure 4). The 1H NMR indicates three different tBu groups, in agreement with a Ci symmetry. The average Cp−Sm distance is 2.55(1) Å, which is longer than the distances reported in the base-free Cpttt2Sm (2.52(1) Å) but smaller than that of the divalent samarium adducts

Figure 1. ORTEP drawing of 1. Thermal ellipsoids are at the 50% probability level; hydrogen atoms have been removed for clarity. Selected bond distances (Å) and angles (deg): Sm−Cp(ctr1), 2.56; Sm−Cp(ctr2), 2.63; Sm−C(3), 3.154; Cp−Sm−Cp, 132.7.

average angle of 163°.25 Similar angles were reported in other samarocene complexes with equatorial coordination.31,33,44 The 1H NMR spectrum of 1 exhibits three resonances at 8.82, 26.0, and −12.7 ppm in a 36:4:2 ratio in agreement with C2v symmetry in solution. This indicates that the intermolecular interaction is not retained in solution at room temperature. The slight paramagnetic shift is in good agreement with the J = 0 ground state of SmII complexes with thermally populated spin− orbit excited states.25,52 Variable-temperature NMR indicates that the complex follows the Curie law over the +20 to −70 °C temperature range since the chemical shift varies linearly versus 1/T (Figure S2 in the Supporting Information). The dissolution of the solvent-free complex 1 in thf followed by crystallization at −35 °C led to the formation of purple crystals of a bis-thf adduct of 1, Cptt2Sm(thf)2 (3) (Figure 2).

Figure 2. ORTEP drawings of 3 and 4. Thermal ellipsoids are at the 50% probability level; hydrogen atoms have been removed for clarity. Selected bond distances (Å) and angles (deg) for 3: Sm−Cp(ctr, av), 2.632(5); Sm−O(av), 2.63(1); Cp−Sm−Cp, 137.9. Selected bond distances (Å) and angles (deg) for 4: Sm−Cp(ctr, av), 2.544(1); Sm− N, 2.691(2); Cp−Sm−Cp, 130.3.

This contrasts with the mono-thf adduct that was reported by Bel’sky in the early 1990s.45 The main metric parameters are gathered in Table 1 and in the Supporting Information. The average Cp−Sm distance is 2.632(5) Å, in good agreement with a SmII−Cp-ring distance. The average Sm−O distance is 2.63(1) Å, which is longer than that reported for the mono-thf adduct of 2.57(1) Å but in agreement with an increased C

DOI: 10.1021/acs.organomet.7b00630 Organometallics XXXX, XXX, XXX−XXX

Article

Organometallics Table 1. Comparison of the Main Structural Parameters of 2−8 (Distances in Å and Angles in deg) Sm−Cp(ctr)

Cptt2Yba

1

3

4

5

6

7

8

2.37

2.56 2.63

2.632(5)

2.55

2.544(1)

2.57(1)

2.692(2)

2.56(2) 2.55(1) 2.358(1)

2.481(8)

2.63(1)

2.539(8) 2.537(4) 2.585(7) 2.558(8)

137.9

132.5

130.3

131.9 130.8

126.6

2.44(1) 2.43(1) 2.07(1) 146.3

4.47

6.18

2.46(1) 2.44(2) 2.48(2) 2.35(1) 2.26(1) 123.4 124.8 4.97

Sm−L

Cp−Sm−Cp

147

132.7

Cptt2Sm(thf)b

Sm−Sm a

From ref 38. bFrom ref 45.

The reaction of both Cptt2Sm (1) and Cpttt2Sm (2) with CO2 (1 atm) has been conducted at low temperature in deuterated toluene. Both reactions are immediate and lead to a discoloration of the solutions with rapid crystallization of pale yellow crystals for the reaction with 1 and pale orange crystals in the case of the reaction with 2. These crystals were suitable for X-ray diffraction and were analyzed at 150 K. They show, in both cases, a bridging carbonate (CO32−) between the two samarocene fragments, [Cpttt2Sm]2(μ-CO3) (7) (Figure 5) and [Cptt2Sm]2(μ-CO3) (8) (Figure 6). Figure 3. ORTEP drawing of 5. Thermal ellipsoids are at the 50% probability level; hydrogen atoms have been removed for clarity. Selected bond distances (Å) and angles (deg). Sm(1)−Cp(ctr, av), 2.539(8); Sm(2)−Cp(ctr, av), 2.537(4); Sm(1)−O(1), 2.585(7); Sm(2)−O(2), 2.558(8); Cp−Sm(1)−Cp, 131.9; Cp−Sm(2)−Cp, 130.8.

Figure 5. ORTEP drawing of 7. Thermal ellipsoids are at the 50% probability level; hydrogen atoms have been removed for clarity. Selected bond distances (Å) and angles (deg): Sm(1)−Cp(ctr, av), 2.481(8); C(35)−O(1), 1.28(4); C(35)−O(2), 1.27(3); C(35)− O(3), 1.29(3); Sm(1)−O(1), 2.44(1); Sm(1)−O(2), 2.43(1); Sm(1)−O(3), 2.07(1); Cp−Sm(1)−Cp, 146.3.

In both 7 and 8 and in agreement with the oxidation of the samarium metal center, the Sm−Cp distances are shorter than these found in divalent samarocene complexes. In 7, the angle is 146.3° and is also longer than those in other compounds and that in 8 (124°). A greater Sm−Sm distance is found (6.18 Å in 7 and 4.97 Å in 8); the lower steric bulk is likely to be the reason for this. The C−O distances in 7 are found within the same range (1.27(3)−1.29(3) Å) and compare well with the carbonate distances (in sodium carbonate; 1.29 Å)57 as well as the three OCO angles of 116, 119, and 124°. The carbonate is disordered over two different positions since there is an inversion center in the molecule and only one crystallographic samarium center. Overall the carbonate binds one samarium in an η2 mode and the other in an η1 mode. The 1H NMR of the crystals of 7 in toluene-d8 showed three signals for the tBu groups at 60 °C, in agreement with an average C2v symmetry in solution. The signals present at 20.0,

Figure 4. ORTEP drawing of 6. Thermal ellipsoids are at the 50% probability level; hydrogen atoms have been removed for clarity. Selected bond distances (Å) and angles (deg): Sm(1)−Cp(ctr, av), 2.56(2); Sm(2)−Cp(ctr, av), 2.55(1); O(1)−O(2), 1.519(6); Sm(1)− O, 2.358(1); Cp−Sm(1)−Cp, 126.6.

reported with this ligand (2.60(2) Å).25 This is in agreement with oxidized samarium metal centers and therefore a peroxo (O22−)-bridged ligand. The O−O distance of 1.519(6) Å is in good agreement with the only other example of this nature reported with samarium (1.538 Å),.54 Moreover, the O−O distance found in the superoxo adduct of Tp′2Sm reported by Takats of 1.319 Å also validates the peroxo proposition.55 Finally, the IR spectrum showed a band at 801 cm−1, in agreement with an O22− group.54,56 D

DOI: 10.1021/acs.organomet.7b00630 Organometallics XXXX, XXX, XXX−XXX

Article

Organometallics

electron reductive disproportionation. Several mechanisms may be proposed, and they have been compared using quantum computations.24 For example, one way considers the formation of an intermediate bridging oxo species and another is concerted. Since 8 does look like a CO2 adduct on an oxo dimer, a mechanistic study is useful. Moreover, at this stage, the solid-state structures do not allow us to quantify the other possible reaction products such as oxalate or polycarbonates formed during the reaction. These considerations will be discussed in the next section. Experimental Mechanistic Studies. Both reactions of CO2 with 1 and 2 were followed by 13C NMR spectroscopy using enriched 13CO2. With 2, the reaction is fast and the appearance of 13CO is immediate at 184.5 ppm, while two broad peaks appear in the carbon spectrum. The signal at 199.1 ppm was attributed to the carbonate species, which indicates a rapid formation of 7 from 2. The other species at 213.3 ppm evolves with time and was not assigned (Figure S11 in the Supporting Information). The reaction of 2 with N2O leads to the clean formation of a new product that crystallized to yield pale yellow crystals that were too small for X-ray diffraction. The 1H NMR is shown is Figure S16 in the Supporting Information and agrees with a symmetrical C2v structure that is in agreement with the formation of the complex [Cpttt2Sm]2(μ-O) (9). The exposure of 9 to 1 atm of CO2 at room temperature does not allow the formation of 7, which indicates that the mechanism does not likely proceed with the oxo intermediate in this case. The reaction with 13CO2 and 1 is less straightforward. The 13 C NMR (as well as the 1H NMR) shows several compounds formed upon the discoloration of the toluene-d8 solution. The formation of 13CO is small in comparison to that observed with 2. In particular the observed broad peak at 123.7 ppm, close to the free 13CO2 chemical shift, might be attributed to the coordinated CO2 or to the bridging CO22− intermediate (see Theoretical Studies and Discussion). The formation of 13CO was also monitored by the appearance of a peak at 184.5 ppm. The NMR data of the crystalline material indicated the presence of a broad peak at 204.3 ppm, which agrees with the presence of the carbonate species. The reaction of 1 with N2O leads also to the clean formation of a pale microcrystalline solid. At first, we were surprized to observe several different products in the 1H NMR (Figure S18 in the Supporting Information). Doing the same reaction starting with 5 instead of 1 showed the same reactivity and the same formation of a pale microcrystalline powder, but the 1H NMR is clean with only one product present (Figure S17 in the Supporting Information), which is in agreement with the formation of [Cpttt2Sm(thf)]2(μ-O) (10). From this 1H NMR it was possible to identify the other product present in the initial NMR; the base-free adduct is extremely sensitive and may pick up any trace of solvents present in the glovebox, leading to the formation of several solvent adducts [Cpttt2Sm(S)]2(μ-O) (S = solvent) having different chemical shifts. This is possible with the Cptt ligand because of the low steric hindrance, allowing coordination of solvent molecules. The addition of CO2 on 10 in toluene-d8 leads to the formation of the carbonate complex 8 (Figure S19 in the Supporting Information). This is in contrast with the absence of a reaction between CO2 and 9. This means that the oxo pathway may be possible for this complex. In order to discriminate between the different pathways for the formation of the carbonate species and in order to

Figure 6. ORTEP drawing of 8. Thermal ellipsoids are at the 50% probability level; hydrogen atoms have been removed for clarity. Selected bond distances (Å) and angles (deg). Sm(1)−Cp(ctr, av), 2.47(1); Sm(2)−Cp(ctr, av), 2.44(2); Sm(1)−O(1), 2.49(1); Sm(2)− O(3), 2.35(1); Sm(1)−O(2), 2.36(1); Sm(2)−O(1), 2.48(1); Cp− Sm(1)−Cp, 123.7; Cp−Sm(2)−Cp, 125.0.

1.06, and −4.13 ppm at 60 °C with a 8:72:36 ratio broaden when the temperature decreases. The variable-temperature (VT) plots of the chemical shift vs 1/T are shown in the Supporting Information. A decoalescence is observed between 280 and 288 K with a free rotation barrier estimated at 12.0(5) kcal mol−1 by the Streiwieser method.25,58 The equivalent tBu and Cp protons split. This is high in comparison to the barrier reported earlier25 and agrees with the dimeric form of 7 that causes larger steric bulk. At lower temperature, other splittings of the tBu group are observed in agreement with the blocking of their free rotation due to steric bulk. In 8, the structural motif is somewhat different. The coordination of the CO32− group is μ-η2-η2 to each samarium atom so that there is a nearly linear coordination between the two samarium atoms and one oxygen atom of the carbonate (O1). The carbon of the carbonate is not well-defined in the structure since the thermal ellipsoid is rather flat, and therefore the C−O carbonate distances are not similar, with two short distances of 1.15(1) and 1.18(1) Å and a longer distance of 1.57(2) Å. Despite our efforts, we were not able to get a better crystallographic model. This structure is very similar to that reported by Gardiner of a carbonate complex with a porphirinogen samarium,11 although the Sm−O1 distances are significantly longer in 8 (Sm(1)−O(1), 2.49(1) Å; Sm(2)− O(1), 2.48(1) Å). The C−O distances in the carbonate reported by Gardiner were also slightly different (1.317(7) and 1.276(4) Å). The 1H NMR of 8 shows three signals at 18.7, 14.4, and −1.83 ppm in a 4:8:72 ratio and agrees with a C2v symmetry, with the C2 axis along the C53−O1 bond. The variabletemperature spectra show a decoalescence temperature located around 210 K (−63 °C). The analysis of the VT NMR allows calculating a free rotation barrier of 8.4(5) kcal mol−1. The lowest steric hindrance in comparison to that for 7 is clearly evidenced by the 3.6 kcal mol−1 energy difference in the rotational barrier. The nature of the coordination situation found in 7 and 8 and the obvious difference in the steric hindrance beget the question of the mechanism for the formation of the carbonate. Formally, the transformation of CO2 into a carbonate is a twoE

DOI: 10.1021/acs.organomet.7b00630 Organometallics XXXX, XXX, XXX−XXX

Article

Organometallics understand why the oxalate species is not observed, although it is the sole product with Cp*2Sm(thf)2,10 we turned to DFT calculations, the object of the next section. Theoretical Studies and Discussion. The first question that was intriguing to us regarded the formation of the peroxo species. Although the nature of the peroxo was rather clear from the structural data, the NMR spectra of both the CObridged complex and the O2-bridged complex would have the same symmetry and both structures were therefore optimized in order to calculate their relative energies. As anticipated, the peroxo-bridged complex is more stable than the CO-bridged adduct by −92.3 kcal mol−1 (Figure 7).

Figure 9. Energy profile of the oxalate and carbonate pathways for the complex Cptt2Sm (1).

electrophilic attack and the discrimination is done at this step. In the case of the bulkier ligand Cpttt, the transition states are 13.8 and 5.5 kcal mol−1 for the C−C attack and the C−O attack, respectively. The difference of 8.3 kcal mol−1 rationalizes the selective formation of the carbonate species with this ligand versus that of the oxalate. In the Cp*2Sm(thf)2 complex, these two TSs were closer in energy (2.5 kcal mol−1).24 The thermodynamic stability of the oxalate therefore rationalizes its formation in 90% yield. Both TSs are kinetically accessible at room temperature, but the formation of the oxalate is easier. Surprisingly, with the less bulky ligand Cptt, the difference in the TSs between the two pathways is identical with that for the Cpttt ligand (8.3 kcal mol−1), which explains the similar outcomes with both ligands. However, with this ligand the TSs are lower in energy. The calculation reproduced well the different geometries of the carbonate species, showing that they do not come from a different mechanism but rather from the hindrance of the ligand. This steric difference is also seen by the possible coordination of a solvent molecule in the oxo-bridged dimer. Finally, the oxo pathway has been calculated with the Cptt ligand, since the experiments determined that the addition of CO2 on the oxo dimer was possible at room temperature. In this case, the TS for the oxo formation is at 21.4 kcal mol−1, greatly higher that the TS for electrophilic attack and its formation is endothermic by 12.7 kcal mol−1, implying that it is the preferred pathway. However, if formed, the oxo readily reacts with CO2 with a very low barrier of 6.2 kcal mol−1, in line with the experiments (see the Supporting Information). The key in the differentiation between the two products (carbonate and oxalate) is therefore related to the two different TSs, the C−C coupling and the C−O coupling. As clearly expressed in this work, the steric hindrance is not the only possible rationalization, since Cptt and Cpttt ligands have different steric bulks; rather, for the oxalate formation, there is a need to localize a negative charge at the carbon in the key intermediate.

Figure 7. Relative energy of [Cpttt2Sm]2(μ-CO) vs 6.

The experimental results showed the formation of the carbonate complexes 7 and 8, and no oxalate-bridged product has been isolated. Moreover, the observed formation of CO is in agreement with the reductive disproportionation of CO2 to yield CO. However, whereas the steric hindrances between Cp* and Cptt are not very different, it is much larger with Cpttt. Therefore, in order to quantify its role on the relative energies, the two different reaction pathwaysleading to either the carbonate or the oxalate producthave been calculated for the reaction of CO2 with 1 and 2 (Figures 8 and 9).

Figure 8. Energy profile of oxalate and the carbonate pathways for the complex Cpttt2Sm (2).

As discussed in earlier work,24 the first step is the formation of a bent CO22− bridging between two samarium fragments, often named the “key intermediate”. The rest of the calculations are based on this intermediate. After this, an electrophilic attack of a second CO2 molecule is involved. This addition involves either the negatively charged oxygen of the CO2 or the carbon atom, leading to the differentiation between the oxalate and the carbonate species. The latter is obtained after several rearrangements and elimination of CO. In both cases the thermodynamically most stable product is the oxalate dimer by 34.1 and 29.7 kcal mol−1 for 2 and 1, respectively, in comparison to the carbonate species. However, the key to the mechanism lies in the transition state (TS) of the



CONCLUSION In conclusion, we have successfully synthesized a series of novel complexes containing samarium. The reaction of CO2 with the base-free complexes Cptt2Sm (1) and Cpttt2Sm (2) leads to the clean formation of dimeric carbonate complexes having different geometrical arrangements in the solid state. The reaction with CO does not lead to any reaction. The extreme F

DOI: 10.1021/acs.organomet.7b00630 Organometallics XXXX, XXX, XXX−XXX

Article

Organometallics

thaw cycles, before being reacted with an atmosphere of CO. Immediately a discoloration of the solution to light yellow was observed. After one night at 17 °C, the reaction was complete and the main product crashed out as orange blocks. 1 H NMR (300 MHz, tol-d8, 294 K): 21.6 (s, 4 H), 21.0 (s, 4 H), 3.33 (s, 36 H), −1.40 (36 H, tBu), 2,82 (s, 36H, tBu), −6.66 (s, 36H, t Bu). Synthesis of [Cpttt2Sm]2(μ-CO3) (7). A 54 mg portion (87.5 μmol) of Cpttt2Sm was dissolved in 2 mL of toluene-d8. The obtained black solution was transferred into a J. Young tapered NMR tube and degassed by three freeze−pump−thaw cycles, before being reacted with an atmosphere of CO2. Immediately a discoloration of the solution to a deep orange color was observed. After one night at 17 °C, the reaction was complete and the main product crashed out as orange blocks (30.4 mg, 54%). 1 H NMR (300 MHz, tol-d8, 333 K) −4.14 (s, 36H, tBu), 1.05 (s, 72H, tBu) 19.95 (bs, 8H, H Cpttt). 13C NMR (75.43 MHz, tol-d8, 298 K) 25.14 (s, tBu), 199.14 (s, CO3). Anal. Calcd for C69H116O3Sm2: C, 64.03; H, 9.03. Found: C, 63.92; H, 9.00. Synthesis of [Cptt2Sm]2(μ-CO3) (8). A 45.5 mg portion (90.1 μmol) of Cptt2Sm (1) was dissolved in 2 mL of toluene-d8. The obtained black solution was transferred into a J. Young tapered NMR tube and degassed by three freeze−pump−thaw cycles, before being reacted with an atmosphere of CO2. Immediately a discoloration of the solution to light yellow was observed. After one night at 17 °C, the reaction was complete and the main product crashed out as a pale yellow microcrystalline solid (13.4 mg, 27.8%). Recrystallization of this solid from pentane at −35 °C afforded crystals suitable for X-ray study. 1 H NMR (300 MHz, tol-d8, 293 K): −1.83 (s, 72H, tBu), 14.43 (s, 8H, H eq Cptt), 18.66 (s, 4H, H Cptt). 13C NMR (75.43 MHz, tol-d8, 293 K): 204.31 (s, CO3). Anal. Calcd for C53H84O3Sm2: C, 59.49; H, 7.91. Found: C, 57.91; H, 7.34. Synthesis of [Cpttt2Sm]2(μ-O) (9). A 52 mg portion (84.3 μmol) of Cpttt2Sm was dissolved in 2 mL of toluene-d8. The obtained black solution was transferred into a J. Young tapered NMR tube and degassed by three freeze−pump−thaw cycles, before being reacted with an atmosphere of N2O. Immediately a discoloration of the solution to light yellow was observed. After one night at 17 °C, the reaction was complete and the title product crashed out as a pale yellow microcrystalline solid (4.7 mg, 5%). 1 H NMR (300 MHz, tol-d8, 293 K): −5.94 (s, 36 H, tBu), −2.41 (s, 36H, tBu), 2.89 (s, 36H, tBu), 20.33 (s, 4H, H Cptt), 22.80 (s, 4H, H Cptt). Synthesis of [Cptt2Sm(thf)]2(μ-O) (10). A 73.4 mg portion (0.126 mmol) of Cptt2Sm (1) was dissolved in 2 mL of toluene-d8. The obtained black solution was transferred into a J. Young tapered NMR tube and degassed by three freeze−pump−thaw cycles, before being reacted with an atmosphere of N2O. Immediately a discoloration of the solution to yellow was observed. After one night at 17 °C, the reaction was complete and the title product crashed out as a yellow microcrystalline solid (13.2 mg, 9%). 1 H NMR (300 MHz, tol-d8, 293 K): −2.85 (s, 72 H, tBu), −1.41 (s, 18 H, thf), 16.83 (s, 8H, H Cptt), 19.74 (s, 4 H, H Cptt). Reactivity of [Cptt2Sm(thf)]2(μ-O) with CO2. A 13.2 mg portion (11.2 μmol) of [Cptt2Sm(thf)]2(μ-O) was dissolved in 2 mL of toluene-d8. The obtained yellow solution was transferred into a J. Young tapered NMR tube and degassed by three freeze−pump−thaw cycles, before being reacted with an atmosphere of CO2. Monitoring of the reaction by 1H NMR showed that the starting material had completely reacted in less than 30 min to yield [Cptt2Sm]2(μ-CO3) (8). X-ray Diffraction. Single crystals of the compounds 1 and 3−8 were mounted on a Kapton loop using Paratone oil and cooled to 150 K (or 260 K for 8) under a nitrogen stream for X-ray structure determination. The loop was transferred to a Nonius Kappa diffractometer using a Mo Kα (λ = 0.71073 Å) X-ray source, a graphite monochromator, and a Bruker APEX-II detector. Data were integrated and corrected for Lorentz and polarization effects. The crystal structures were solved in SHELXS59 and SHELXT60 and refined in SHELXL61 by full-matrix least squares using anisotropic

care that is needed in these studies is highlighted by the serendipitous reaction of traces of O2 and Cpttt2Sm leading to the clean formation of an original bridged side-on peroxo complex: [Cpttt2Sm]2(μ-O2) (6). The mechanism for the formation of the carbonate species in place of the possible oxalate species (that was reported with Cp*2Sm) was investigated from experimental and theoretical perspectives and led to the conclusion that all reaction pathways are rather close in energy. Therefore, the design of both the sterics and the electronics of divalent lanthanide complexes for selective small-molecule activation is very important, since a subtle change can result in different possible mechanisms leading to different products.



EXPERIMENTAL SECTION

General Considerations. All reactions were performed using standard Schlenk-line techniques or in an argon-filled glovebox (MBraun). All glassware was dried at 120 °C for at least 12 h prior to use. KCptt, Cptt2Sm(thf) (5), and Cpttt2Sm were prepared according to published procedures. thf, Et2O, pyridine, toluene-d8, thf-d8, and pentane were dried over sodium and transferred under reduced pressure in a cold flask. All solvents were degassed prior to use. CO2, CO, and N2O were purchased from Air Liquide as Alphagaz N48. Labeled CO2 was purchased from Eurisotop. NMR spectra were recorded in 5 mm tubes adapted with a J. Young valve on Bruker 300 MHz Avance III spectrometers. Chemical shifts are expressed relative to TMS in ppm. Elemental analyses were obtained from Mikroanalytisches Labor Pascher. Cpttt2Sm was synthesized as reported earlier.25 Syntheses. Synthesis of Cptt2Sm (1). A 50 mL portion of Et2O was added to a mixture of SmI2 (2.01 g, 4.97 mmol, 1 equiv) and KCptt (2.15 g, 9.94 mmol, 2 equiv). The obtained brownish green slurry was stirred overnight before solids were removed by filtration, and the solvent was removed in vacuo, yielding 2.42 g (84%) of Cptt2Sm(OEt2) as a brownish green powder. This complex was then dissolved in a minimum of toluene (10 mL), and the brown solution was transferred to a tube equipped with a greaseless J. Young tap and containing a magnetic stir bar. The flask was opened to dynamic vacuum briefly, closed, and then heated to 70 °C. The solvent was removed under dynamic vacuum over a period of 2 h, and the black residue was redissolved in toluene (10 mL); this procedure was repeated two times. Finally the black residue was extracted with pentane and dried under reduced pressure to yield the title product as a black powder (1.78 g, 71%), which can be recrystallized, if necessary, in pentane or in toluene. 1 H NMR (300 MHz, tol-d8, 294 K): 8.82 (s, 36 H, tBu), 25.99 (s, 4 H, H Cptt), −12.73 (s, 2H, H Cptt). Anal. Calcd for C26H42Sm: C, 61.84; H, 8.38. Found: C, 55.82; H, 8.21. The elemental analysis is low for carbon despite our efforts. Synthesis of Cptt2Sm(thf)2 (3). A 71.0 mg portion (0.141 mmol) of Cptt2Sm (1) was dissolved in 0.5 mL of thf. Storage at −35 °C of the obtained purple solution yielded the product as large purple crystals (34.7 mg, 38%). 1 H NMR (300 MHz, tol-d8, 294 K): 14.28 (s, 36 H, tBu), 14.28 (s, 4 H, H Cptt), −13.6 (s, 2 H, H Cptt), −4.02 (s, 4H, thf), −4.55 (s, 4H, thf). Due to easy desolvation, the elemental analysis was not satisfactory. Synthesis of Cptt2Sm(py) (4). A 58.9 mg portion (0.117 mmol) of Cptt2Sm (1) was dissolved in 1 mL of thf. Storage at −35 °C of the obtained black solution yielded the product as large black crystals (30.0 mg, 44%). 1 H NMR (300 MHz, tol-d8, 294 K): 15.03 (s, 36H, tBu), 13.15 (s, 4H), −9.97 (s, 2H), 3.13 (2H, py), 2.82 (1H, py), −0.93 (2H, py). Anal. Calcd for C31H47NSm: C, 63.75; H, 8.11; N, 2.40. Found: C, 62.66; H, 8.00; N, 2.98. Isolation of [Cpttt2Sm]2(μ-O2) (6). Cpttt2Sm was dissolved in 2 mL of toluene-d8. The obtained black solution was transferred into a J. Young tapered NMR tube that was degassed by three freeze−pump− G

DOI: 10.1021/acs.organomet.7b00630 Organometallics XXXX, XXX, XXX−XXX

Article

Organometallics

(3) Tolman, W. B. Angew. Chem., Int. Ed. 2010, 49, 1018. (4) Labinger, J. A.; Bercaw, J. E. Nature 2002, 417, 507. (5) Gunsalus, N. J.; Koppaka, A.; Park, S. H.; Bischof, S. M.; Hashiguchi, B. G.; Periana, R. A. Chem. Rev. 2017, 117, 8521. (6) Summerscales, O. T.; Cloke, F. G. N.; Hitchcock, P. B.; Green, J. C.; Hazari, N. Science 2006, 311, 829. (7) Franke, R.; Selent, D.; Börner, A. Chem. Rev. 2012, 112, 5675. (8) Liu, Q.; Wu, L.; Jackstell, R.; Beller, M. Nat. Commun. 2015, 6, 5933. (9) Engel, E. R.; Jouaiti, A.; Bezuidenhout, C. X.; Hosseini, M. W.; Barbour, L. J. Angew. Chem., Int. Ed. 2017, 56, 8874. (10) Evans, W. J.; Seibel, C. A.; Ziller, J. W. Inorg. Chem. 1998, 37, 770. (11) Davies, N. W.; Frey, A. S. P.; Gardiner, M. G.; Wang, J. Chem. Commun. 2006, 4853. (12) Evans, W. J.; Perotti, J. M.; Brady, J. C.; Ziller, J. W. J. Am. Chem. Soc. 2003, 125, 5204. (13) Andrez, J.; Pécaut, J.; Bayle, P.-A.; Mazzanti, M. Angew. Chem., Int. Ed. 2014, 53, 10448. (14) Tsoureas, N.; Castro, L.; Kilpatrick, A. F. R.; Cloke, F. G. N.; Maron, L. Chem. Sci. 2014, 5, 3777. (15) Formanuik, A.; Ortu, F.; Inman, C. J.; Kerridge, A.; Castro, L.; Maron, L.; Mills, D. P. Chem. - Eur. J. 2016, 22, 17976. (16) Inman, C. J.; Frey, A. S. P.; Kilpatrick, A. F. R.; Cloke, F. G. N.; Roe, S. M. Organometallics 2017, DOI: 10.1021/acs.organomet.7b00263. (17) Cooper, O.; Camp, C.; Pécaut, J.; Kefalidis, C. E.; Maron, L.; Gambarelli, S.; Mazzanti, M. J. Am. Chem. Soc. 2014, 136, 6716. (18) Castro-Rodriguez, I.; Meyer, K. J. Am. Chem. Soc. 2005, 127, 11242. (19) Castro-Rodriguez, I.; Nakai, H.; Zakharov, L. N.; Rheingold, A. L.; Meyer, K. Science 2004, 305, 1757. (20) Evans, W. J.; Grate, J. W.; Bloom, I.; Hunter, W. E.; Atwood, J. L. J. Am. Chem. Soc. 1985, 107, 405. (21) Lam, O. P.; Franke, S. M.; Heinemann, F. W.; Meyer, K. J. Am. Chem. Soc. 2012, 134, 16877. (22) Schoo, C.; Bestgen, S.; Schmidt, M.; Konchenko, S. N.; Scheer, M.; Roesky, P. W. Chem. Commun. 2016, 52, 13217. (23) Louyriac, E.; Roesky, P. W.; Maron, L. Dalton Trans 2017, 46, 7660. (24) Castro, L.; Labouille, S.; Kindra, D. R.; Ziller, J. W.; Nief, F.; Evans, W. J.; Maron, L. Chem. - Eur. J. 2012, 18, 7886. (25) Nocton, G.; Ricard, L. Dalton Trans 2014, 43, 4380. (26) Riemschneider, R.; Nehring, R. Monatsh. Chem. 1959, 90, 568. (27) Leigh, T. J. Chem. Soc. 1964, 0, 3294. (28) Dehmlow, E. V.; Bollmann, C. Z. Naturforsch., B: J. Chem. Sci. 1993, 48, 457. (29) Venier, C. G.; Casserly, E. W. J. Am. Chem. Soc. 1990, 112, 2808. (30) Fleischmann, M.; Jones, J. S.; Gabbai, F. P.; Scheer, M. Chem. Sci. 2015, 6, 132. (31) Arleth, N.; Gamer, M. T.; Köppe, R.; Konchenko, S. N.; Fleischmann, M.; Scheer, M.; Roesky, P. W. Angew. Chem., Int. Ed. 2016, 55, 1557. (32) Scherer, O. J.; Werner, B.; Heckmann, G.; Wolmershäuser, G. Angew. Chem., Int. Ed. Engl. 1991, 30, 553. (33) Gun’ko, Y. K.; Bulychev, B. M.; Soloveichik, G. L.; Belsky, V. K. J. Organomet. Chem. 1992, 424, 289. (34) Sofield, C. D.; Andersen, R. A. J. Organomet. Chem. 1995, 501, 271. (35) Ortu, F.; Fowler, J. M.; Burton, M.; Formanuik, A.; Mills, D. P. New J. Chem. 2015, 39, 7633. (36) Schultz, M.; Boncella, J. M.; Berg, D. J.; Tilley, T. D.; Andersen, R. A. Organometallics 2002, 21, 460. (37) Schultz, M.; Burns, C. J.; Schwartz, D. J.; Andersen, R. A. Organometallics 2001, 20, 5690. (38) Schultz, M.; Burns, C. J.; Schwartz, D. J.; Andersen, R. A. Organometallics 2000, 19, 781. (39) Yu, X.-Y.; Jin, G.-X.; Hu, N.-H.; Weng, L.-H. Organometallics 2002, 21, 5540.

thermal displacement parameters for all non-hydrogen atoms. The positions of the hydrogen atoms were determined using residual electronic densities, which were calculated by a difference Fourier procedure. Finally, in order to obtain a complete refinement, a weighting step followed by multiple loops of refinement was done. Details on crystal data and structure refinements are summarized in Tables S3 and S4 in the Supporting Information. ORTEP drawings were produced using Mercury. CIF files are deposited at the Cambridge Data Base Centre under the reference CCDC numbers 1567712−1567718 for 1 and 3−8, respectively. Theoretical Computations. Calculations were performed with the Gaussian 09 program62 at the DFT level of theory using the hybrid functional B3PW91.63,64 Samarium was treated with a large-core Stuttgart−Dresden relativistic effective core potential (RECP), adapted to the +3 oxidation state, used in combination with its optimized basis set augmented by a set of f polarization functions (α = 1.000).65 Hydrogen, oxygen, carbon, and sulfur atoms were described with a 6-31G+(d,p) double-ζ-quality basis set. Electronic energies and enthalpies were computed at T = 298 K in the gas phase. Geometry optimizations were performed without any symmetry constraints, and analytical frequency calculations allowed verification of the nature of the extrema. Intrinsic reaction coordinate (IRC) calculations were carried out to verify the connections of the optimized transition states. Natural bond orbital (NBO) analysis66 was used to analyze electron density.



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.organomet.7b00630. NMR data, IR data, and details of X-ray crystallography and NMR studies (PDF) Cartesian coordinates of the calculated structures (XYZ) Accession Codes

CCDC 1567712−1567718 contain the supplementary crystallographic data for this paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by emailing [email protected], or by contacting The Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033.



AUTHOR INFORMATION

Corresponding Author

*E-mail for G.N.: [email protected]. ORCID

Carine Clavaguéra: 0000-0001-5531-2333 Laurent Maron: 0000-0003-2653-8557 Grégory Nocton: 0000-0003-0599-1176 Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This project was financed by the French National Agency with grant number ANR-15-CE29-0019. We thank the CNRS and Ecole polytechnique for funding. M.X. is grateful to the DGA for funding. L.M. is a member of the Institut Universitaire de France.



REFERENCES

(1) Crutchley, R. J. Coord. Chem. Rev. 2017, 334, 1. (2) Milani, B.; Licini, G.; Clot, E.; Albrecht, M. Dalton Trans. 2016, 45, 14419. H

DOI: 10.1021/acs.organomet.7b00630 Organometallics XXXX, XXX, XXX−XXX

Article

Organometallics (40) Schultz, M.; Sofield, C. D.; Walter, M. D.; Andersen, R. A. New J. Chem. 2005, 29, 919. (41) Schumann, H.; Glanz, M.; Winterfeld, J.; Hemling, H.; Kuhn, N.; Kratz, T. Chem. Ber. 1994, 127, 2369. (42) Jaroschik, F.; Nief, F.; Le Goff, X. F.; Ricard, L. Organometallics 2007, 26, 3552. (43) Nief, F.; de Borms, B. T.; Ricard, L.; Carmichael, D. Eur. J. Inorg. Chem. 2005, 2005, 637. (44) Gun’ko, Y. K.; Bulychev, B. M.; Belsky, V. K.; Soloveichik, G. L. J. Organomet. Chem. 1992, 440, 47. (45) Bel’sky, V. K.; Gunko, Y. K.; Bulychev, B. M.; Sizov, A. I.; Soloveichik, G. L. J. Organomet. Chem. 1990, 390, 35. (46) Ren, W.; Song, H.; Zi, G.; Walter, M. D. Dalton Trans.s 2012, 41, 5965. (47) Ren, W.; Zi, G.; Fang, D.-C.; Walter, M. D. Chem. - Eur. J. 2011, 17, 12669. (48) Ren, W.; Lukens, W. W.; Zi, G.; Maron, L.; Walter, M. D. Chem. Sci. 2013, 4, 1168. (49) Ren, W.; Zhao, N.; Chen, L.; Zi, G. Inorg. Chem. Commun. 2013, 30, 26. (50) Lukens, W. W.; Beshouri, S. M.; Blosch, L. L.; Stuart, A. L.; Andersen, R. A. Organometallics 1999, 18, 1235. (51) Schumann, H.; Winterfeld, J.; Köhn, R. D.; Esser, L.; Dietrich, A.; Sun, J. Chem. Ber. 1993, 126, 907. (52) Van Vleck, J. H. The Theory of Electric and Magnetic Susceptibilities; Oxford University Press: London, 1932. (53) Khvostov, A. V.; Bulychev, B. M.; Belsky, V. K.; Sizov, A. I. Russ. Chem. Bull. 1999, 48, 2162. (54) Neumüller, B.; Weller, F.; Gröb, T.; Dehnicke, K. Z. Anorg. Allg. Chem. 2002, 628, 2365. (55) Zhang, X.; Loppnow, G. R.; McDonald, R.; Takats, J. J. Am. Chem. Soc. 1995, 117, 7828. (56) Holland, P. L. Dalton Trans. 2010, 39, 5415. (57) Dusek, M.; Chapuis, G.; Meyer, M.; Petricek, V. Acta Crystallogr., Sect. B: Struct. Sci. 2003, 59, 337. (58) Luke, W. D.; Streitwieser, A. J. Am. Chem. Soc. 1981, 103, 3241. (59) Sheldrick, G. Acta Crystallogr., Sect. A: Found. Crystallogr. 2008, 64, 112. (60) Sheldrick, G. Acta Crystallogr., Sect. A: Found. Adv. 2015, 71, 3. (61) Sheldrick, G. Acta Crystallogr., Sect. C: Struct. Chem. 2015, 71, 3. (62) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Barone, V.; Mennucci, B.; Petersson, G. A., et al. Gaussian 09 Revision D.01, Gaussian Inc., Wallingford, CT, 2009. (63) Burke, K.; Perdew, J. P.; Wang, Y. Electronic Density Functional Theory: Recent Progress and New Directions; Plenum: New York, 1998. (64) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (65) Dolg, M.; Stoll, H.; Savin, A.; Preuss, H. Theor. Chim. Acta 1989, 75, 173. (66) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988, 88, 899.

I

DOI: 10.1021/acs.organomet.7b00630 Organometallics XXXX, XXX, XXX−XXX